α-Methylacyl-CoA racemase (AMACR)
![AMACR as a Drug Target for Prostate Cancer
The enzyme α-methylacyl-CoA racemase (AMACR) is consistently overexpressed in prostate, colon, breast and other cancers and its overexpression correlates with increased enzymatic activity. It is widely used as an immunohistochemical biomarker to confirm diagnosis for prostate cancer and has recently been identified by using siRNA and shRNA knockdown of the AMACR gene as a potential therapeutic target.
AMACR catalyses the epimersation of branched chain α-methylacyl-CoA esters such as pristanoyl-CoA 1 (Fig.1), the bile acid intermediate trihydroxycoprostanoyl CoA (THC-CoA) and some profens such as ibuprofen CoA and fenoprofen-CoA. Pristanoyl-CoA is derived from the diet either directly or by α-oxidation of dietary phytanic acid and exists as (2R)- and (2S)-stereoisomers. The AMACR interconverts (R)- and (S)-stereoisomers, providing the (S)-isomer for the stereospecific branched-chain acyl-CoA oxidase in the first step of β-oxidation. The AMACR enzyme is a cofactor independent racemase and the mechanism involves a 1,1-proton transfer, thought to proceed through an enzyme-bound enolate (Fig. 2).

Figure 1: AMACR epimersies the α-stereochemistry of branched chain fatty acids to allow them to enter β-oxidation

Figure 2: Proposed two-base mechanism for AMACR showing reversible deprotonation at the alpha carbon centre

Fig. 3: Acyl CoA substrate bound to MCR, the bacterial homologue of AMACR.
People with a congenital absence of AMACR suffer late onset neuropathy as a result of a build up of unmetabolised (R)-pristanoyl CoA. However, this can be managed by excluding phytol from the diet so there is an opportunity to develop drugs to target AMACR as a treatment for cancer. We have been designing and synthesizing AMACR inhibitors as structural probes and potential drug molecules (Figure 4).[1] Our initial series of compounds were substrate analogues containing one or more fluorine atoms adjacent to the α-centre. These substrate analogues were initially designed to be suicide inhibitors by β-elimination of fluorine to generate a reactive Michael acceptor within the active site. However, the compounds turned out to be competitive inhibitors against the natural substrate THC-CoA with Ki values in the range 1μM - 20μM.

Figure 4: Competitive Inhibitors of AMACR
We also tested ibuprofen-CoA as single isomers. Interestingly the (R)-isomer was 4-fold more potent. Clearly the CoA renders these inhibitors non-bioavailable. However we have shown that the precursor acid of compound 1 is thioesterified by a cell free extract in the presence of CoA and that it shows dose-dependent cancer cell growth inhibition that corresponds to AMACR expression levels. We are currently developing more potent inhibitors using a rational approach involving structural studies. One key challenge is to design inhibitors that do not require the coenzyme A component.
[1] Carnell, A. J.; Hale, I.; Denis, S.; Wanders, R. J. A.; Isaacs, W. B.; Wilson, B. A.; Ferdinandusse, S. J. Med. Chem. 2007, 50, 2700–2707.](Medicinal%20Chemistry_files/shapeimage_2.png)
